Detection and isolation of faulty measurements in medical Wireless Sensor Networks

In this paper, we propose a new framework for online detection and isolation of faulty measurements reported by medical wireless sensors. In our proposed framework, each sensor applies the non-seasonal Holt-Winter algorithm to detect any deviation in the time series associated with its measurements, and it will only transmit the detected abnormal values to the portable collection device (smart phone), in order to reduce the consumed energy of data transmission. As the physiological parameters are heavily correlated, the faulty measurements are usually uncorrelated with values of other sensors. Therefore, the collection device uses this correlation and its global view on the number of received deviations, to decide whether to raise an alarm for an emergency situation or to discard reported faulty measurements. Our main objective is to reduce false alarms triggered by faulty measurements.We apply our proposed approach on real physiological data set, and we prove its ability to achieve good detection accuracy with a low false alarm rate.